It is known that a singlet oxygen molecule is generated from the chemical reaction of chlorine gas with a mixed solution of hydrogen peroxide solution (H2O2) and potassium hydroxide (KOH) or sodium hydroxide (NaOH). A chemical oxygen iodine laser (generally called COIL, COIL being an abbreviation for Chemical Oxygen Iodine Laser) which operates as a laser by transferring energy of O2(1Δg) to an iodine atom (I) is commonly known as a high energy laser of 1.315 um. Non-Patent literature 1 to 4 explains about the chemical oxygen iodine laser.
One of the reasons for utilizing the energy transfer from the excited O2(1Δg) to iodine is that it is considered that direct lasing from O2(1Δg) is difficult. Actually, there has been no report concerning direct lasing from O2(1Δg). However, there is a report which says weak light was detected in an experiment which aims at the direct lasing of O2(1Δg). In the experiment, a spectrum observation was not performed. Non-Patent literature 5 is the only a report which says that the direct lasing from O2(1Δg) was successful. According to the Non-Patent literature 5, although the lasing of the oxide laser was confirmed, it was a small amount of energy. Since there has been no other report concerning the direct lasing of O2(1Δg), it has been considered that the realization of an oxygen laser is difficult. It is considered that one of the reasons for this is that the long spontaneous emission lifetime of O2(1Δg) makes the laser gain, which is inversely proportional to the spontaneous emission lifetime, quite small.
It has been reported that the spontaneous emission lifetime is approximately 72 mins. which is orders of magnitude longer than that of other lasers. For example, the spontaneous emission lifetime of the excited iodine is approximately 130 ms, which is shorter than that of O2(1Δg) by four digits. Also, YAG laser widely used as a solid laser has a spontaneous emission lifetime of approximately 230 μs (the radiation lifetime for the laser transition is approximately 550 μs), which is shorter than that of O2(1Δg) by seven digits. The spontaneous emission lifetime for O2(1Δg) is further described in a technical paper such as a paper of Non-Patent literature 6.
Meanwhile, the stimulate emission coefficient (B) which influences the easiness of lasing is expressed as the relation (1). Since the stimulate emission coefficient (B) is proportional to the spontaneous emission coefficient (A), which is the reciprocal of the spontaneous emission lifetime, the stimulate emission coefficient for O2(1Δg) becomes quite small. This results in a small gain.[Equation 1]A21/B21=8πv213h/c3  (1)
Where h is the Plank constant, v21 is a frequency of the transition, and c is the speed of light.
However, the small gain does not mean that lasing is impossible, but simply means that lasing is difficult. Since a long gain length is considered to oscillate, simulations are performed to confirm this oscillation.
The simulations are based on rate equations which are typically used to estimate laser power. The utilized rate equations are listed in the table 1. A spectral profile g (v) shown in the following equation 2 is used for the simulations which consider both the Doppler broadening and the pressure broadening.
                              [                      Equation            ⁢                                                  ⁢            2                    ]                ⁢                                                                                                g          ⁡                      (            v            )                          =                  2          ⁢                                                    ln                ⁢                                                                  ⁢                2                            π                                ⁢                                    V              ⁡                              (                                  a                  ,                  x                                )                                                    Δ              ⁢                                                          ⁢                              v                D                                                                        (        2        )            
Where V (a, x) is the Voigt profile which can be calculated as the following equation (3). For example, the following Non-Patent literature 7 explains this point.
                              [                      Equation            ⁢                                                  ⁢            3                    ]                ⁢                                                                                                V          ⁡                      (                          a              ,              x                        )                          =                              a            π                    ⁢                                    ∫                              -                ∞                            ∞                        ⁢                                                            ⅇ                                      -                                          y                      2                                                                                                            a                    2                                    +                                                            (                                              x                        -                        y                                            )                                        2                                                              ⁢                                                          ⁢                              ⅆ                y                                                                        (        3        )            
Where a and x are variable numbers defined by the following equations (4) and (5).
                              [                      Equation            ⁢                                                  ⁢            4                    ]                ⁢                                                                                      a        =                                            ln              ⁢                                                          ⁢              2                                ⁢                                    Δ              ⁢                                                          ⁢                              v                c                                                    Δ              ⁢                                                          ⁢                              v                D                                                                        (        4        )                                          [                      Equation            ⁢                                                  ⁢            5                    ]                ⁢                                                                                      x        =                  2          ⁢                                    ln              ⁢                                                          ⁢              2                                ⁢                                    (                              v                -                                  v                  0                                            )                                      Δ              ⁢                                                          ⁢                              v                D                                                                        (        5        )            
A calculation result of the spectral profile factor g(v) is shown in FIG. 7 with the parameter of the O2 pressure. Since the profile has a curved shape against a frequency v, g (v) value depending on the divided frequency should be considered in the simulation in order to accurately estimate the laser output energy/power. However, to confirm whether lasing is possible, the peak value of the g (v) is used to save computation time.
When considering the laser oscillation of the oxide laser, vibrational levels and rotational levels for the oxygen molecule should be considered. According to Non-Patent literature 8, since most oxygen molecules populate the lowest vibrational level of V=1, the simulations consider the population distribution of only rotational levels to save computation time. The population density N (j) of each rotational level of oxygen molecules can be calculated by the following equation (6).
                              [                      Equation            ⁢                                                  ⁢            6                    ]                ⁢                                                                                                N          ⁡                      (            J            )                          =                  N          ⁢                                    hcB              v                                                      k                B                            ⁢              T                                ⁢                      (                                          2                ⁢                                                                  ⁢                J                            +              1                        )                    ⁢                      exp            ⁡                          [                              -                                                      E                    ⁡                                          (                      J                      )                                                                                                  k                      B                                        ⁢                    T                                                              ]                                                          (        6        )            
Where kB is a Boltzmann constant, and By is the rotational constant with the value of 42.5 GHz. The By is further explained in the following Non-Patent literature 8. N (J) becomes a carved line shown in FIG. 8 in the case of T=273 deg in the equation (6).
However, the population density N (J) illustrated in FIG. 8 is the sum of the population densities for the three branches which are classified in accordance with the difference between the J number of the upper level and that of the lower level. A branch is called an Q branch when J of the upper level is the same as that of the lower level (therefore ΔJ=0), a branch is called an P branch when J of the lower level is one greater than that of the upper level (therefore, ΔJ=−1), and a branch is called an R branch when J of the lower level is one smaller than that of the upper level (therefore, ΔJ=+1). The population in consideration of transition probabilities of the P, Q, and R branches is illustrated in FIG. 9. It should be noted that, among the three branches, the Q branch gives the strongest emission. Non-Patent literature 10 explains about the transition probability of each branch. The Non-Patent literature 1 reports a measurement result in which the spontaneous emission from the Q branch is the strongest one among the three branches.
Since the lasing starts at the transition from the level with the largest population, the peak value of the population curve is taken into the consideration in the simulation. Although it is considered that the laser transition can simultaneously occur at the several lines with their gains higher than the oscillation threshold, the simulation considers only the peak values of the Q branches.
A computer program of the simulation as explained above has been developed based on the rate equations in consideration of the reaction equations listed in the table 1 in FIG. 10. A space along the laser cavity is divided into a plurality of areas as shown in FIG. 11, and the population density and photon density for each level for each small area are used as variables in the rate equation. In the simulation of the laser cavity, two variables are used for the laser photons, one being for the photons travelling in one direction, the other being for the photons travelling in the opposite direction. However, the first variable is used for the laser amplifier simulation.
The conditions for calculating the simulations are listed in the table 2 in FIG. 12. An excited oxygen generation efficiency of 60% is assumed, considering typical singlet oxygen generators. This means 60% of generated oxygen molecules are assumed to be O2(1Δg), while 40% of the same are O2(3Σg). A total oxygen pressure of 1000 Pa (˜0.1 atm) is assumed considering some high pressure singlet oxygen generators. An oxygen temperature of 0 degrees C. (273.15K) is assumed. And the H2O pressure is assumed to be equal to the saturation vapor pressure at 0 degrees C.
The configuration of the oxygen molecule laser oscillator shown in FIG. 6 is simulated. A laser cavity 615 has an output mirror 607 and a total reflector 608. The output mirror 607 and the total reflector 608 are installed inside a housing 601. Before generating excited oxygen molecules in order to fill the inside of the laser cavity 615 with the excited oxygen molecules, the laser cavity 615 is evacuated (the pumping direction is expressed as an arrow 603). After the evacuation, a valve 604 is closed in order not to pump out the excited oxygen molecules and to fill the inside of the laser cavity 615 with the excited oxygen molecules.
Since it is necessary not to oscillate until the inside of the laser cavity 615 is filled with enough excited oxygen molecules, the total reflector 608 needs to be off-aligned so that the radiation generated inside the laser cavity 615 will not be resonated. Therefore the total reflector 608 is a rotating mirror which is rotated so that the total reflector 608 and the output mirror 607 become parallel only when the laser oscillates. That is, the total reflector 608 as the rotating mirror functions as a Q switch. Such a rotating-mirror Q switch is described in the following Non-Patent Literature 9.
At first, in order to estimate a minimum required length of the laser cavity, an 02 laser amplifier is simulated. The simulation results in a small-signal gain coefficient against a gain length of the amplifier are shown in FIG. 13. As illustrated in FIG. 13, a gain length of more than 7 m is necessary to obtain a gain of more than 1%.
Therefore, to be able to oscillate, the O2 laser oscillator is simulated, assuming a cavity length of 10 m, an output mirror reflectivity of 99.0% and a total reflector reflectivity of 99.9%. The simulation result is shown in FIG. 14. The horizontal axis denotes an elapsed time from a Q switched operation after filling the inside of the laser cavity with the excited oxygen molecules, and the vertical axis denotes time variation of the output laser power density. The result indicates that lasing can be possible with a pulse half width of 20-30 μs at the time of ˜0.15 ms. Since the plotted output power density is only for a single rotational transition, a much larger amount of output energy can be expected considering all the possible laser transitions.
As mentioned above, the simulation shows the possibility of lasing by lengthening a gain length so that it is a long one of 10 m. To realize an actual device with such a long gain length, a large number of excited oxygen molecules are necessary, which might require a big high-flow-rate singlet oxygen generator. However, a big singlet oxygen generator is not necessarily required considering that the singlet oxygen has quite a long radiation lifetime, which allows sufficient time for generating enough singlet oxygen molecules. This is one of the advantages of oxygen molecule lasers.